Gas combustion turbines are used for a variety of applications such as driving an electric generator in a power generating plant or propelling a ship or an aircraft. Firing temperatures in modern gas turbine engines continue to increase in response to the demand for higher efficiency engines. Superalloy materials have been developed to withstand the corrosive high temperature environment that exists within a gas turbine engine. However, even superalloy materials are not able to withstand extended exposure to the hot combustion gas of a current generation gas turbine engine without some form of cooling and/or thermal insulation.
Thermal barrier coatings are widely used for protecting various hot gas path components of a gas turbine engine. The reliability of such coatings is critical to the overall reliability of the machine. The design limits of such coatings are primarily determined by laboratory data. However, validation of thermal barrier coating behavior when subjected to the stresses and temperatures of the actual gas turbine environment is essential for a better understanding of the coating limitations. Such real world operating environment data is very difficult to obtain, particularly for components that move during the operation of the engine, such as the rotating blades of the turbine.
Despite the extreme sophistication of modern turbine engines, such as gas turbines for generating electrical power or aircraft engines for commercial and military use, designers and operators have very little information regarding the internal status of the turbine engine components during operation. This is due to the harsh operating conditions, which have prevented the use of traditional sensors for collecting reliable information of critical engine components.
Many current turbines are equipped with sensors capable of limited functions such as exhaust gas-path temperature measurements, flame detection and basic turbine operating conditions. Based on this information, turbine operators such as utility companies operate engines in a passive mode, in which maintenance is scheduled based on prior histories of similar engines. Engine rebuilds and routine maintenance are performed in the absence of a prior knowledge of the remaining or already utilized life of individual components. The lack of specific component information makes early failure detection very difficult, often with the consequence of catastrophic engine failure due to abrupt part failure. This results in inefficient utilization, unnecessary downtime and an enormous increase in operating cost.
Currently, the gas turbine industry approach is to depend on the measurement of gas path temperature, which is related back to specific component problems based on experience and history regarding a class of engines. This approach is highly subjective and only allows for determining already severe situations with an engine. It does not provide indications of impending damage or insight into the progression of events leading up to and causing engine damage due to component degradation or failure.
The instrumentation of a component such as a blade or vane within a steam turbine typically includes placing wire leads on the balance wheel, which continue on to the blade airfoil. The wire leads are typically held together by an epoxy. These wires are routed from within the component to the turbine casing. The pressure boundary of a component may be breached to introduce a sensor such as a thermocouple and a braze is back filled to hold the thermocouple in place. Each thermocouple sensor has wire leads coming out of the component that are connected back to a diagnostic unit. Instrumenting a plurality of components of a turbine in this manner results in an extensive network of wires just for monitoring the single operating condition of temperature. Instrumenting components using this technique is expensive, which is a barrier to instrumenting a large number of components within a single turbine. Further, the wire leads and data transfer is frequently poor, which can result in costly repairs and flawed data analysis.
Using thermocouples for temperature measurements in the gas path of a turbine may be disadvantageous because it only provides feedback to an operator that a temperature change has occurred in the gas path. It does not provide any indication as to why the temperature change has occurred. For diagnosing problems with blades or vanes based on a measured temperature change, there has to be an historical correlation between the measured temperature differential and the specific problem, such as a hole in a vane. This correlation is difficult and time consuming to derive to within a reasonable degree of certainty and needs to be done on an engine-by-engine basis taking into account turbine operation conditions. When a temperature differential is measured, it is difficult, if not impossible, to predict what the problem is or where it is located. Consequently, the turbine must typically be shut down and inspected to determine the scope of repair, replacement or other maintenance to be performed.
In any application, combustion turbines are routinely subject to various maintenance procedures as part of their normal operation. Diagnostic monitoring systems for gas turbines commonly include performance monitoring equipment that collects relevant trend and fault data used for diagnostic trending. In diagnostic trend analysis, certain process data (such as exhaust gas temperature, fuel flow, rotor speed and the like) that are indicative of overall gas turbine performance and/or condition are compared to a parametric baseline for the gas turbine. Any divergence of the raw trend data from the parametric baseline may be indicative of a present or future condition that requires maintenance. Such diagnostic monitoring systems can only predict or estimate specific component conditions and do not collect data from or provide any analysis with respect to the actual condition of a specific component itself.
In this respect, conventional methods of predicting component failure for gas turbines and of scheduling maintenance have not been entirely accurate or optimized. The traditional “duty cycle” used for predictive maintenance does not reflect real operational conditions, especially off-design operations. The actual life of specific components of a gas turbine depends strongly on the actual usage of that gas turbine and the specific components within the turbine.
For example, elevated temperatures and stresses within the turbine, and aggressive environmental conditions may cause excessive wear on components in the turbine beyond that predicted with the standard design duty cycle. Off-design operating conditions, which are often experienced by industrial gas turbines, are not reflected by the standard duty cycles. The actual life of components in the gas turbine may be substantially less than that predicted by the design duty cycle. Alternatively, if more favorable conditions are experienced by an actual gas turbine than are reflected in the design duty cycle, the actual component life may last substantially longer than that predicted by maintenance schedules based on the design duty cycle. In either event, the standard design duty cycle model for predicting preventive maintenance does not reliably indicate the actual wear and tear experienced by gas turbine components.
Known techniques for predicting maintenance and component replacement rely on skilled technicians to acquire or interpret data regarding the operation of a combustion turbine. Such techniques are subject to varying interpretations of that data by technicians. Technicians may manually evaluate the operational logs and/or data collected from gas turbines. Technicians, for example, may evaluate start and stop times and power settings to determine how many duty cycles had been experienced by the gas turbine, their frequency, period and other factors. In addition, if the data log of a gas turbine indicated that extraordinary conditions existed, such as excessive temperatures or stresses, the technicians may apply “maintenance factors” to quantify the severity of these off-design operational conditions.
None of these techniques provide accurate information with respect to the actual condition of a specific component or component coating, which may lead to unnecessary repair, replacement or maintenance being performed causing a significant increase in operating costs.